Abstract
Seafloor resource development in the future is expected to be accompanied by the mining of metal ores, as these sulfide ores contain valuable metals. However, the mining process is detrimental to marine oceanic environments. As a precautionary approach, innovations in the collection of environmental baseline data and new onboard assessment methods for marine environmental impacts are required. Due to the low cell density in open ocean water, techniques for rapid impact assessment of seawater without sample concentration are required. The purpose of this study was to establish a heavy metal impact assessment system for seawater samples of marine microbes using a portable flow cytometer, On-chip Sort. We established a protocol for detecting heavy metal-induced damage to cells via propidium iodide (PI) staining using algal culture strains (Bathycoccus prasinos NIES-2670, Synechococcus sp. NIES-969, Prochlorococcus sp. NIES-2885, and Cyanobium sp. NIES-981) obtained from the marine environment maintained at the Microbial Culture Collection of the National Institute for Environmental Studies, Tsukuba, Japan. Results showed that PI staining could detect the effects of heavy metals on cells. The proportion of PI-positive cells increased with an increase in the concentration of heavy metal mixture or copper exposure. Compared to cyanobacteria, damaged cells of eukaryotic algae were detected. Therefore, the effects of heavy metals on both eukaryotic and prokaryotic algae can be rapidly assessed via PI-based flow cytometry using samples containing low cell densities.
Commercial mining operations are planned at seafloor sulfide deposits to obtain base metals such as copper (Cu), lead (Pb), zinc (Zn), and rare metals (Hoagland et al. 2010, Petersen et al. 2016). However, the unavoidable contamination of seawater by these metals during mining operations represents a major concern (Van Dover 2011, Boschen et al. 2013, Wedding et al. 2015, Hauton et al. 2017). Environmental impact assessments, baseline monitoring, mitigation, and management are important for identifying and reducing the potential impacts of deep-sea mining (Jones et al. 2019). A recent study reported that hydrothermal sulfide ore particulates represent a source of toxic pollutants that can affect the primary production of plankton (Fuchida et al. 2017). Assessment of the leaching of hazardous elements from ores into seawater before the commencement of full-scale operations is necessary to minimize the environmental impacts of deep-sea mining operations (Fuchida et al. 2017).
Cu, Pb, Zn, and arsenic can be selectively released from sulfide minerals with different chemical compositions present in hydrothermal fields (Fuchida et al. 2017). Fuchida et al. (2018) found that heavy metals, such as Zn and Pb, can leach from sulfide samples at an accelerated rate in the surface environment [high temperate (20°C) and aerobic] rather than in the seabed environment [cold (5°C) and anaerobic conditions] (Fuchida et al. 2018), and their recent experiment has revealed the galvanic interaction could be a primal reason for the selective release of Zn and Pb from hydrothermal ores (Fuchida et al. 2021). In addition to Zn and Pb, the presence of Cu at low concentrations and picomolar levels can affect the number of Prochlorococcus (Mann et al. 2002). In deep-sea environments, biodiversity loss and potential compensatory actions associated with deep-sea mining have also been identified (Niner et al. 2018). To enable scientific evidence-based action, it will be necessary to establish a method for in situ seawater quality monitoring and environmental impact assessment methods. A pilot bioassay has recently been performed. For example, a delayed fluorescence (DF)-based bioassay was performed using the Cyanobium sp. NIES-981 to evaluate the toxicity of core samples obtained from three drill holes at the Izena Hole, middle Okinawa Trough, East China Sea collected during the Japan Agency for Marine-Earth Science and Technology (JAMSTEC) drilling vessel (D/V) Chikyu cruise. They found that the DF inhibition was a result of high zinc and lead concentrations in the leaches (Yamagishi et al. 2018).
There are various limitations when performing toxicity assessments on board. Only a few research vessels can carry large scientific instruments such as incubators and mass spectrometers. We have developed a system to efficiently evaluate toxicity in a small space, such as a research vessel. The DF assay is one such bioassay system that enables the evaluation of toxicity in a shorter period than that required for completing growth assays (Katsumata et al. 2006, Yamagishi et al. 2016). Although this method is simple and rapid compared to the growth assay, it requires a high concentration of cells to measure DF (approximately ＞107 cells mL−1 of Cyanobium sp. cells). A direct impact assessment using on-site seawater would be preferable, seafloor mining areas are often oligotrophic environments, and microbial cells in seawater are very sparse, making it impossible to use on-site seawater directly instead of concentrated samples. Since the concentration process may cause various types of damage to cells, there is a need for an alternative system that uses seawater directly.
Flow cytometry (FCM) can detect the fluorescence or light scattered by individual cells via a high-throughput process. The FCM approach may clarify the composition and interactions within microbial systems in oceans (van den Engh et al. 2019). However, FCM alone cannot be used to detect cell damage and requires additional steps of cell staining. Additionally, for on-site experiments, a flow cytometer is required to be compact and portable for facilitating installation on a research vessel. On-chip Sort is a benchtop-type and disposable microfluidic chip-based flow cytometer/cell sorter. The disposable microfluidic chip allows sorting and analysis in damage-, contamination-, and maintenance-free manner, and has been used in medical research and related areas (Watanabe et al. 2014, Sawada et al. 2016, Hasegawa et al. 2019). Here, we used this system to evaluate heavy metal sensitivity in marine phytoplankton, even in low-cell density cultures.
The marine cyanobacteria Prochlorococcus and Synechococcus are abundant in oceanic environments. The analysis of Prochlorococcus containing low chlorophyll content using a benchtop flow cytometer underestimates cell counts due to inadequate separation of the autofluorescence signal from background noise (Gérikas Ribeiro et al. 2016). One solution involves staining the nucleic acids before FCM analysis (Marie et al. 1997). SYBR Green I (SYBR) is often used to stain nucleic acids in both live and dead cells and can be detected using a blue laser (488 nm). Membrane-permeable nucleic acid staining dyes are used to distinguish between living and dead or damaged cells. PI is widely used as a fluorescent probe to assess cell viability based on membrane integrity. Since PI is membrane-impermeable, cells are considered to contain disrupted membranes (dead or damaged cells) if their nuclei are stained with PI. Previous applications include studies of bacteria in the river and marine environments (Grégori et al. 2001, Nescerecka et al. 2016), and microalgal cultures (Gumbo et al. 2014, Nogueira et al. 2015, Suman et al. 2015).
Although several studies have used PI-based assays for algae, few studies have used PI-based assays to evaluate heavy metal-induced damage in both marine eukaryotic and prokaryotic plankton. We developed a PI-based FCM assay system using algal cultures to assess the effects of heavy metals. This study targeted four strains from marine environments: a green alga Bathycoccus prasinos NIES-2670 was selected as a representative of eukaryotic algae which are sensitive to Cu, Zn, and Pb (Ota et al. 2020); Synechococcus sp. NIES-969 and Prochlorococcus sp. NIES-2885 were selected as representative oceanic cyanobacteria; Cyanobium sp. NIES-981 was selected since it is used as a marine test organism for toxicological risk assessments (Alquezar and Anastasi 2013, Yamagishi et al. 2016).
Materials and methods
Culture conditionsAlgal strains were obtained from the Microbial Culture Collection at the National Institute of Environmental Studies (MCC-NIES), Tsukuba, Japan (http://mcc.nies.go.jp). Four strains were used in this study: Cyanobium sp. NIES-981, Synechococcus sp. NIES-969, Prochlorococcus sp. NIES-2885, and B. prasinos NIES-2670 (Fig. 1, Table 1). Cells were pre-cultured at 20°C under light at 20–40 µmol photons m−2 s−1 with a 12 : 12-h light (L) : dark (D) cycle in a glass test tube containing 10 mL of ESM medium for NIES-969, NIES-981, and NIES-2670, (Carlos et al. 1999) or Pro-99 medium for NIES-2885 (Moore et al. 2007).
Table 1. List of strains examined in this study.
| Strain No. | Species | Class | Phylum | Major diameter |
|---|
| NIES-969 | Synechococcus sp. | Cyanophyceae | Cyanophyta | 1–2 µm |
| NIES-981 | Cyanobium sp. | Cyanophyceae | Cyanophyta | 1–3 µm |
| NIES-2670 | Bathycoccus prasinos | Prasinophyceae | Chlorophyta | 2 µm |
| NIES-2885 | Prochlorococcus sp. | Cyanophyceae | Cyanophyta | 1 µm |
Cells were observed using a Nikon ECLIPSE Ni-U microscope (Nikon, Tokyo, Japan) containing a differential interference contrast optic, and images were obtained using a DS-Fi3 digital camera (Nikon, Tokyo, Japan). To observe the staining process, live and dead cells were stained with a final concentration of 0.03 mM PI (Dojindo, Kumamoto, Japan). Heat-treated control cells (dead cells) were prepared by boiling cultures for a few seconds (ca. 100°C). After staining cells with PI for 1.5 min, time-lapse microscopy was performed using a DFC450C digital camera mounted on a DMi8 inverted microscope (Leica Microsystems, Wetzlar, Germany). PI-stained cells were observed using an RHOD fluorescent cube (excitation: 546/12 nm, emission: 585/40 nm). The images were obtained over 16 min with a capture interval of 1.5 min. Red channel images (512×512 pixels) were saved as stacked images. Twelve cells in the stack images were randomly selected, and the region of interest (ROI) was set for each cell.
PI staining-based FCM assaySolutions containing either Cu or a mixture of metals were used as stress sources. Artificial mixed-metal solutions (CK) were prepared concerning the leaching experiment data of the JAMSTEC D/V Chikyu CK16-05 cruise in 2016 (Core sample ID number: C9027B 1X-CC), in which Mn, Fe, Zn, As, Cd, Sb, and Pb concentrations were 164, 72.6, 14,700, 33.7, 22.0, 37.9, and 2,420 ppb, respectively (Yamagishi et al. 2018). Cultures at the late logarithmic growth to stationary phase were diluted with sterile seawater to eliminate chelation effects in the medium (final concentration 1×106 cells mL−1). Cu exposure was performed by adding a CuSO4 stock solution (100 mg L−1) to the samples. For CK exposure, a concentrated stock solution (10×CK stock solution) was added to the culture. After Cu or CK addition, samples were incubated in a 48-well microplate (1 mL/well) for 24 h at 20°C under illumination (L : D=12 : 12) at 60 µmol photons m−2 s−1 using white light-emitting diodes.
After incubation, 1 µL of 1 mg mL−1 PI was added to 200 µL of the sample. The samples were incubated for at least 15 min at room temperature (25–28°C) in the dark. SYBR Green I (SYBR) (BioWhittaker, Rockland, ME, USA) was added to the culture at a final concentration of 1 : 10,000 (commercial stock). Immediately before FCM analysis, the stained culture was diluted 4–5 times with filtered seawater to reduce the background dye (centrifugation was not possible in this experimental system). The samples were then analyzed using an On-chip Sort flow cytometer (On-Chip Biotechnologies, Tokyo, Japan) using 488 nm blue laser excitation. The number of cells analyzed for NIES-2670 and cyanobacterial strains was 400 and 2,000 cells, respectively. The following detection channels were used: FL2 (543/22 nm), FL3 (591.5/43 nm), and FL5 (716/40 nm) (Table 2).
Table 2. Detection filters and fluorescent dye/autofluorescence used in this study.
| Detection filter | Detection wavelength | Color | Fluorescent dye or autofluorescence |
|---|
| FL2 | 543/22 nm | Green | SYBR Green I, Phycoerythrin |
| FL3 | 591.5/43 nm | Yellow/orange | Propidium iodide |
| FL5 | 716/40 nm | Red | Chlorophyll a |
For time-lapse imaging analysis, the intensity of light in the ROI was quantified using the Plot Z-axis Profile of ImageJ 1.51 s (https://imagej.nih.gov/ij). A flow cytometry dot plot and histogram analysis were performed using OnchipFlow ver. 1.7.9.0. (http://on-chipbio.com). Significant differences were evaluated using the Student’s t-test. p-values＜0.01 were considered significant.
Results
Separation of noise and signalTo confirm whether the On-chip Sort used in this study is appropriate for studying the three oceanic phytoplankton, B. prasinos (NIES-2670), Synechococcus sp. (NIES-969), and Prochlorococcus sp. (NIES-2885), we evaluated the gating for the separation of background noise and signal, and the detection filters to be used (Table 2). Usually, forward scatter (FSC; cell size) as well as autofluorescence of chlorophyll (red) or phycoerythrin (PE) (yellow/orange) can be analyzed to separate signals from background noise. When filtered seawater without cells is analyzed via FCM, only background noise was presented in the two-dimensional (2D) plot (Fig. 2A, left and middle). No signal was plotted in this case (Fig. 2A, right). In eukaryotic algae (NIES-2670), red fluorescence separated the signal from noise due to the presence of a sufficient amount of chlorophyll (Fig. 2B, left). Green fluorescence could not separate the signal from the noise, but FSC could separate them (Fig. 2B, middle). Here, we decided to obtain the signal by gating with red fluorescence (Fig. 2B, right). In NIES-969 containing PE, both red and green fluorescence were analyzed to separate the signal from the noise (Fig. 2C, left and middle). Here, we decided to obtain the signal by gating with green fluorescence (Fig. 2C, right).
In unstained NIES-2885 cells, the signal and noise could not be separated by either red or green fluorescence, or by FSC (Fig. 2D). This may be due to the low chlorophyll content and absence of PE in NIES-2885. Moreover, the signals and noise could not be separated in PI-stained cells (Fig. 2E). In contrast, when SYBR staining was performed, the signal of green fluorescence could be separated (Fig. 2F, middle), and the signal could be obtained by gating using green fluorescence (Fig. 2F, right).
Validation experimentTo determine whether PI staining can identify damaged cells, validation experiments were performed using live and dead cells of three representative marine algal strains (NIES-2670, -969, and -2885) and the test strain NIES-981 (Table 1). Cyanobium sp. NIES-981 is a PE-free cyanobacterium that has been used as a marine test strain (Yamagishi et al. 2016, 2018, Ota et al. 2020). In addition, the whole genome of this strain has been sequenced (Yamaguchi et al. 2016), and the strain is used as a model alga in ecotoxicology studies. First, freeze-thawed cells were used as damaged control cells, considering that phycobiliproteins such as PE are heat sensitive. Additionally, heat-treated (boiled) cells were also used because cyanobacteria could be resistant to freezing.
FCM analysis was performed using frozen and live cells along with PI-stained and unstained cells (Fig. 3). Among the unstained cells, PI-positive cells accounted for less than 1% of both frozen and live cells. The proportion of PI-stained cells was 4.19% among live cells and 16.5% among frozen cells of NIES-2670 (Fig. 3A). Similarly, the proportion of PI-positive NIES-981 cells increased from 0.264 to 5.98% (Fig. 3B). The validation experiment showed that PI fluorescence intensity differed between live and frozen cells, and the proportion of PI-positive cells increased among frozen cells of both algae and cyanobacteria.
As mentioned above, since cyanobacteria are prokaryotes, they possess a simple cell structure and a higher tolerance to freezing than eukaryotic cells. When we compared the PI fluorescence values of NIES-969 and NIES-2885 between frozen and live cells, there was no difference in PI fluorescence (Fig. S1, S2). We performed heat-treatment control experiments using PE-free NIES-981 (Fig. S3A) and NIES-2885 (Fig. S3B). In NIES-981 cultures, PI-positive cells (FL3 ＞40 a.u.) were detected only among heat-treated cells stained with PI, but not among other cells (Fig. 3A). In contrast, a comparison of live and heat-treated cells stained only with SYBR showed that heat treatment caused changes in the FL3 fluorescence intensity in NIES-2885 (Fig. S3B). Due to the remarkable effect of heat treatment on Prochlorococcus, care should be taken in the interpretation of the heat treatment experiment results.
To evaluate the efficiency of PI staining from another aspect, NIES-981 cells stained with PI were observed using time-lapse fluorescence microscopy (Fig. 4). Quantification of red fluorescence intensity in each cell in the sequential images showed that the fluorescence intensity of heat-treated cells was significantly higher than that of living cells and that the fluorescence intensity increased over time (Fig. 4C). This confirmed that PI was only taken up by dead cells, leading to an increase in fluorescence intensity. Considering that PI fluorescence saturates at approximately 15 min of exposure to the stain (Fig. 4C), the PI assays were performed for at least 15 min of staining.
Metal exposure assaysMetal exposure PI-based assays were performed using four strains: NIES-2670, NIES-969, NIES-2885, and NIES-981. Exposure tests using CK for 24 h are shown in Fig. 5. In the alga NIES-2670, the proportion of PI-positive cells increased as the metal concentration increased, reaching 81.9% upon exposure to 100% CK (equivalent to the metal concentration eluted in the test) (Fig. 5A). It should also be noted that chlorophyll fluorescence (FL5) of NIES-2670 was reduced in cultures exposed to CK concentrations of 50% and 100%. Chlorophyll fluorescence was observed up to 1K intensity (a.u. in FL5) when cells were exposed to CK concentrations below 20%; however, cells were concentrated at almost 100 fluorescence intensity (a.u. in FL5) when exposed to CK concentration above 50% (Fig. 5A). This may be due to the chlorophyll-quenching effect at high metal concentrations.
PE-containing NIES-969 and SYBR-stained NIES-2885 showed green to yellow fluorescence. Since the PE or SYBR emission fluorescence partially overlaps with PI fluorescence, the threshold cannot be set at a region where positive and negative cells are distinguished. In this study, we set an arbitrary threshold at a position shifted to the right from the peak of the histogram (for example, NIES-969; Fig. S4, right histogram), and observed the proportion of cells showing increased PI fluorescence. We confirmed that the FL3 fluorescence value increased, indicating that damaged cells could be detected. In NIES-981, a minor increase in the proportion of PI-positive cells was observed in cultures exposed to CK concentrations of 50% and 100% (Fig. 5D).
After setting arbitrary thresholds as described above and analyzing cells via FCM, we confirmed that the FL3 fluorescence of metal-exposed cells shifted in the direction of increase compared to that of non-exposed cells. When comparing cultures exposed to 0% and 50% CK, the number of cells showing increased fluorescence increased from 22.4 to 63.2% for NIES-969 (Fig. 5B) and from 11.5 to 26.1% for NIES-2885 (Fig. 5C).
The Cu solution exposure tests performed for 24 h are shown in Fig. 6. Similar to CK exposure results, the proportion of PI-positive cells increased in Cu exposure in NIES-2670 cultures; PI-positive cells were observed at a Cu concentration of 0.01 mg mL−1, and the proportion of positive cells reached 39.8% at 1 mg mL−1concentration (Fig. 6A). Notably, the chlorophyll fluorescence (FL5) of NIES-2670 was reduced in cultures exposed to Cu concentrations of 0.1–1 mg mL−1. This quenching may be due to the same mechanism as in the case of CK exposure. Metal-exposed cells of NIES-969 and NIES-2885 were similarly evaluated via FCM. Compared to non-exposed cells, metal-exposed cells showed an upward shift in FL3 fluorescence. When comparing cultures exposed to 0 mg mL−1 and 0.1 mg mL−1 Cu2+, the proportion of cells showing increased fluorescence increased from 23.5 to 31.3% for NIES-969 and from 10.4 to 23.3% for NIES-2885 (Fig. 6B, C). In NIES-981 cultures, a minor increase in the proportion of PI-positive cells was observed in cultures exposed to Cu concentrations of 1 mg mL−1 (Fig. 6D).
Discussion
Deep-sea mining for acquiring metal ores is expected to damage marine environments, and new onboard assessment methods for marine environmental impacts are required. At present, onboard techniques for assessing environmental damage to marine microbes face several challenges. Therefore, we developed an assay to evaluate the impact of metal toxicity on phytoplankton using seawater directly. Before seawater analysis, we used oceanic strains of cyanobacteria and green algae from MCC-NIES. The use of cultured strains instead of natural seawater allows us to perform robust and controlled tests for the optimization of FCM. A previous study showed that B. prasinos NIES-2670 is sensitive to Zn, Pb, and Cu (Ota et al. 2020); thus, it is an ideal representative test strain for analysis. Considering that previous studies have applied the PI method to algae (Xiao et al. 2011, Gumbo et al. 2014, Nogueira et al. 2015, Suman et al. 2015, Zhang et al. 2015), and the present validation experiment showed that damaged cells show enhanced yellow to orange fluorescence (Figs. 1, 2), we confirmed that PI-based FCM may be feasible for evaluating cell viability.
Since the FCM approach provides more accurate and high-throughput measurements than those obtained via microscopic observation (Marie et al. 2005), FCM has been widely used to analyze the dynamics of phytoplankton populations in the oceanographic field (Olson et al. 1990, Campbell et al. 1994, Marie et al. 2014). To examine cell viability via FCM, it is necessary to use a fluorescent dye such as PI which evaluates membrane integrity. It should be noted that PI-based assays require fresh cells because fixed or frozen cells cannot be used. Therefore, it is crucial to observe fresh cells on-site, and a portable flow cytometer is needed for this purpose.
In recent years, compact and portable flow cytometers have been developed, facilitating their use in field experiments such as on research vessels. However, different detection sensitivities have been reported for conventional and benchtop flow cytometers. For example, a comparison between the Becton Dickinson (BD) laboratory-based flow cytometry FACSCanto and a benchtop type BD Accuri C6 showed that Prochlorococcus and Synechococcus populations are underestimated using the benchtop device (Gérikas Ribeiro et al. 2016). This disadvantage can be compensated for by staining cells with a fluorescent dye such as SYBR, which can stain both live and dead cells. In this study, we confirmed that the signal could not be separated from the noise using unstained Prochlorococcus cells; in contrast, SYBR staining facilitated separation. However, for cells that do not require SYBR staining, such as eukaryotic cells, PI staining alone is sufficient. In this study, we proposed a gating strategy via FCM using On-chip Sort for major algal groups (Fig. 7).
The results of this study confirmed that PI staining could detect the effects of heavy metals on cells. Compared to cyanobacteria, damaged cells of eukaryotic algae were detected. This may be attributed to the fact that eukaryotic algae contain sufficient amounts of chlorophyll, and their genome size is larger than that of prokaryotes, suggesting that the PI stain signal itself is stronger and detectable. For example, PI-positive cells were detected at 20% of the concentration used in the elution experiment under CK exposure (Fig. 5A). This complex metal solution contains various metals other than copper (Yamagishi et al. 2018), among which Zn and Pb are the major metals.
For the 10% CK exposure experiment, strains NIES-2670, NIES-969, and NIES-2885 were found to show more than a 1% increase in the proportion of PI-positive cells. Given the criterion that positive cells can be considered as positive when at least ten positive cells are present in the gate (2.5% or more when 400 cells are analyzed), the detection limit seems to be around 10% for CK or 0.01 mg mL−1 for Cu exposure except for NIES-981 (Roederer 2008). This suggests that these three strains were detectable for metal damage after at least 10% CK exposure. Similarly, strains NIES-2670, NIES-969, and NIES-2885 were found to show more than a 1% increase in the proportion of PI-positive cells after 0.01 mg mL−1 Cu exposure. This suggests that these three strains can be evaluated for metal damage at 0.01 mg mL−1 Cu. Overall, the PI-based FCM assay may be applied to typical algae present in oceanic waters. In contrast, more than 1% of PI-positive cells were detected among NIES-981 cells exposed to 50% CK or 1 mg mL−1 Cu. The reason for the higher detection limit of NIES-981 compared to that of other species remains unknown. However, we speculate that one of the reasons for this observation is that NIES-981 may show metal tolerance to a certain extent since it is a coastal species.
Two subpopulations were identified in the 2D plot of NIES-2885 (Figs. 5C, 6C). The culture of NIES-2885 was non-axenic, and colorless bacteria were also stained with SYBR, suggesting that these bacteria appeared in the signal gate. However, it is unclear which subpopulation corresponded to these bacteria. These two populations may not be separated based on chlorophyll fluorescence, but rather by the difference in the amount of nucleic acid due to SYBR fluorescence. The spillover fluorescence of SYBR may have been detected by the fluorescent filter of FL5. A large amount of colorless bacteria exists in oceanic water (105–106 cells/mL orders) (Li 1998, Li et al. 2004), and these results indicate that the present assay system can be applied to colorless bacteria.
In addition to PI-based FCM, the DF assay is a promising technique for use in field experiments as an alternative technique to the growth assay based on the OECD test guidelines (Katsumata et al. 2006, Yamagishi et al. 2016). While the OECD-based growth assay requires 72 h of incubation, the DF assay has the advantage of analyzing cells in a few hours to 24 h. In addition, the DF assay requires only a few milliliters of culture volume, whereas the OECD-based growth assay requires 100 mL per culture and thus requires a large incubator. In the DF assay using NIES-981, the detection was possible even at exposure concentrations below 10% CK (Yamagishi et al. 2018), whereas in the PI assay, the detection was possible at CK concentrations above 50%. This suggests that the detection sensitivity of the PI assay was lower than that of the DF assay for the test strain NIES-981. This may be due to intrinsic differences in the targets of observation. The DF assay analyzes photons generated by the excitation of a photosystem reaction center in bulk cells, whereas the PI assay evaluates the degree of PI staining based on the membrane integrity of individual cells.
The advantage of PI-based FCM is that oceanic seawater can be used directly; for the DF assay, a relatively high density of cells is required; for example, a cell density of 107 cells mL−1 or more of NIES-981 is required (Yamagishi et al. 2018). The PI assay proposed in this study can detect cells at a density of 105–106 cells mL−1 or less. At this concentration, detection via the DF assay was below the detection limit (3×105 cells mL−1 for NIES-981; Fig. S5). There are advantages and disadvantages to both DF and FCM assays: DF assay has a higher sensitivity but requires a higher cell density for assays, whereas the PI assay is relatively less sensitive than DF, but can analyze on-site seawater containing low cell densities (＜106 cells mL−1). It will be necessary to construct a rapid and highly sensitive evaluation system for metal toxicity in the future by considering the beneficial features of both methods.
Future studies should consider the following three aspects: first, full-scale demonstrations using natural seawater should be performed. Although we do not include data in this main data, preliminary investigations onboard have suggested that the present system can be applied to natural seawater (Fig. S6). At present, insufficient data have been collected, and additional fieldwork is necessary.
Second, the FCM assay can be used for all major marine plankton. However, it is not necessary to analyze all microbial groups in the field. For example, since eukaryotic algae are usually more abundant in coastal areas than in open oceans, it may be possible to target only eukaryotic algae in coastal areas. In contrast, it may be possible to target Prochlorococcus as the dominant phytoplankton in oligotrophic ocean waters (Fig. 7). The advantage of this assay system is that it can be used for a variety of groups according to the type of dominant phytoplankton in the survey area. Conducting a basic community analysis via FCM in advance will enable the identification of appropriate groups in advance, and consequently, more accurate and detailed environmental assessments can be performed on board.
Third, a limitation includes crosstalk between spillover fluorescence (green fluorescence and PI) when measuring the intensity of PI fluorescence after separating the signal and noise based on green fluorescence. In this study, we determined that cells were damaged if they showed enhanced FL3 fluorescence, and we confirmed that the fluorescence was enhanced via exposure to high metal concentrations. For SYBR-stained or PE-containing cells, an arbitrary threshold was set slightly to the right of the histogram peak (Fig. S4). This threshold setting confirmed the enhancement of PI fluorescence due to metal stress. This crosstalk does not occur in eukaryotic algae; however, it affects the analysis of cyanobacteria showing green to orange fluorescence. One solution involves developing an SYBR -free detection system. However, this depends on the sensitivity of the flow cytometer which is still technically difficult to improve in portable flow cytometers. Recently, studies have reported the use of multicolor lasers to optimize separation (Thompson and van den Engh 2016). Future studies should consider a system optimized for cyanobacteria using multicolor light. On-chip Sort can also be used to detect wavelengths in the near-infrared region (FL6; 775/46 nm). We confirmed that it is possible to separate the signal from the noise using these wavelengths (Fig. S6). Although we did not use this detection filter (FL6) in this study, the use of the near-infrared region may allow us to increase the detection parameters and avoid crosstalk issues.
The mining of seabed resources is expected to progress further. Before the commencement of full-scale deep-sea mining, it is necessary to establish a basis for feasible monitoring and develop environmental assessment techniques onboard. The current study demonstrated an FCM-based method to evaluate the impact of metal toxicity on phytoplankton in pelagic water. In this study, we conducted experiments using representative strains of marine algae and cyanobacteria and succeeded in detecting the toxic effects of metals. The FCM data suggested that the sensitivity of the method enables its use at a practical level for both algae and cyanobacteria. We were able to detect the impact on cyanobacteria based on the current methods; however, future improvements in gating strategy may be necessary for populations comprising PE-containing or SYBR green-stained cells. Further improvements include not only the optimization of the staining method but also updating filter settings and excitation wavelength in FCM analysis.
With Supplementary data of Figures (Figs. S1–S6).
Acknowledgments
We thank Jun Amemura at NIES for technical assistance and the staff of the Microbial Culture Collections of NIES for providing the strains. This work was supported by the Cabinet Office, Government of Japan, Cross-ministerial Strategic Innovation Promotion Program, Next-Generation Technology for Ocean Resource Exploration, and NIES Research Funding type A.
References
- Alquezar, R. and Anastasi, A. 2013. The use of the cyanobacteria, Cyanobium sp., as a suitable organism for toxicity testing by flow cytometry. Bull. Environ. Contam. Toxicol. 90: 684–690.
- Boschen, R. E., Rowden, A. A., Clark, M. R. and Gardner, J. P. A. 2013. Mining of deep-sea seafloor massive sulfides: A review of the deposits, their benthic communities, impacts from mining, regulatory frameworks and management strategies. Ocean Coast. Manage. 84: 54–67.
- Campbell, L., Nolla, H. A. and Vaulot, D. 1994. The importance of Prochlorococcus to community structure in the central North Pacific Ocean. Limnol. Oceanogr. 39: 954–961.
- Carlos, A. A., Baillie, B. K., Kawachi, M. and Maruyama, T. 1999. Phylogenetic position of Symbiodinium (Dinophyceae) isolates from tridacnids (Bivalvia), cardiids (Bivalvia), a sponge (Porifera), a soft coral (Anthozoa), and a free-living strain. J. Phycol. 35: 1054–1062.
- Fuchida, S., Ishibashi, J., Shimada, K., Nozaki, T., Kumagai, H., Kawachi, M., Matsushita, Y. and Koshikawa, H. 2018. Onboard experiment investigating metal leaching of fresh hydrothermal sulfide cores into seawater. Geochem. Trans. 19: 15.
- Fuchida, S., Kawachi, M. and Koshikawa, H. 2021. Kinetic investigation of galvanic dissolution of ZnS and PbS with FeS2 from hydrothermal sulfides in seawater. Appl. Geochem. 129: 104963.
- Fuchida, S., Yokoyama, A., Fukuchi, R., Ishibashi, J. I., Kawagucci, S., Kawachi, M. and Koshikawa, H. 2017. Leaching of metals and metalloids from hydrothermal ore particulates and their effects on marine phytoplankton. ACS Omega 2: 3175–3182.
- Gérikas Ribeiro, C., Marie, D., Lopes dos Santos, A., Pereira Brandini, F. and Vaulot, D. 2016. Estimating microbial populations by flow cytometry: Comparison between instruments. Limnol. Oceanogr. Methods 14: 750–758.
- Grégori, G., Citterio, S., Ghiani, A., Labra, M., Sgorbati, S., Brown, S. and Denis, M. 2001. Resolution of viable and membrane-compromised bacteria in freshwater and marine waters based on analytical flow cytometry and nucleic acid double staining. Appl. Environ. Microbiol. 67: 4662–4670.
- Gumbo, J. R., Cloete, T. E., van Zyl, G. J. J. and Sommerville, J. E. M. 2014. The viability assessment of Microcystis aeruginosa cells after co-culturing with Bacillus mycoides B16 using flow cytometry. Phys. Chem. Earth 72: 24–33.
- Hasegawa, N., Takeda Nakamura, I., Ueno, T., Kojima, S., Kawazu, M., Akaike, K., Okubo, T., Takagi, T., Suehara, Y., Hayashi, T., Saito, T., Kaneko, K., Mano, H. and Kohsaka, S. 2019. Detection of circulating sarcoma tumor cells using a microfluidic chip-type cell sorter. Sci. Rep. 9: 20047.
- Hauton, C. et al. 2017. Identifying toxic impacts of metals potentially released during deep-sea mining—A synthesis of the challenges to quantifying risk. Front. Mar. Sci. 4: 368.
- Hoagland, P., Beaulieu, S., Tivey, M. A., Eggert, R. G., German, C., Glowka, L. and Lin, J. 2010. Deep-sea mining of seafloor massive sulfides. Mar. Policy 34: 728–732.
- Jones, D. O. B., Durden, J. M., Murphy, K., Gjerde, K. M., Gebicka, A., Colaço, A., Morato, T., Cuvelier, D. and Billett, D. S. M. 2019. Existing environmental management approaches relevant to deep-sea mining. Mar. Policy 103: 172–181.
- Katsumata, M., Koike, T., Nishikawa, M., Kazumura, K. and Tsuchiya, H. 2006. Rapid ecotoxicological bioassay using delayed fluorescence in the green alga Pseudokirchneriella subcapitata. Water Res. 40: 3393–3400.
- Li, W. K. W. 1998. Annual average abundance of heterotrophic bacteria and Synechococcus in surface ocean waters. Limnol. Oceanogr. 43: 1746–1753.
- Li, W. K. W., Head, E. J. H. and Glen Harrison, W. 2004. Macroecological limits of heterotrophic bacterial abundance in the ocean. Deep Sea Res. Part I Oceanogr. Res. Pap. 51: 1529–1540.
- Mann, E. L., Ahlgren, N., Moffett, J. W. and Chisholm, S. W. 2002. Copper toxicity and cyanobacteria ecology in the Sargasso Sea. Limnol. Oceanogr. 47: 976–988.
- Marie, D., Partensky, F., Jacquet, S. and Vaulot, D. 1997. Enumeration and cell cycle analysis of natural populations of marine picoplankton by flow cytometry using the nucleic acid stain SYBR Green I. Appl. Environ. Microbiol. 63: 186–193.
- Marie, D., Rigaut-Jalabert, F. and Vaulot, D. 2014. An improved protocol for flow cytometry analysis of phytoplankton cultures and natural samples. Cytometry A 85: 962–968.
- Marie, D., Simon, N. and Vaulot, D. 2005. Phytoplankton Cell Counting by Flow Cytometry. In: Andersen, R. A. (ed.). Algal Culturing Techniques. Academic Press, London. pp. 253–267.
- Moore, L. R., Coe, A., Zinser, E. R., Saito, M. A., Sullivan, M. B., Lindell, D., Frois-Moniz, K., Waterbury, J. and Chisholm, S. W. 2007. Culturing the marine cyanobacterium Prochlorococcus. Limnol. Oceanogr. Methods 5: 353–362.
- Nescerecka, A., Hammes, F. and Juhna, T. 2016. A pipeline for developing and testing staining protocols for flow cytometry, demonstrated with SYBR Green I and propidium iodide viability staining. J. Microbiol. Methods 131: 172–180.
- Niner, H. J., Ardron, J. A., Escobar, E. G., Gianni, M., Jaeckel, A., Jones, D. O. B., Levin, L. A., Smith, C. R., Thiele, T., Turner, P. J., Van Dover, C. L., Watling, L. and Gjerde, K. M. 2018. Deep-sea mining with no net loss of biodiversity—An impossible aim. Front. Mar. Sci. 5: 53.
- Nogueira, P. F. M., Nakabayashi, D. and Zucolotto, V. 2015. The effects of graphene oxide on green algae Raphidocelis subcapitata. Aquat. Toxicol. 166: 29–35.
- Olson, R. J., Chisholm, S. W., Zettler, E. R. and Armbrust, E. V. 1990. Pigments, size, and distributions of Synechococcus in the North Atlantic and Pacific Oceans. Limnol. Oceanogr. 35: 45–58.
- Ota, S., Yamaguchi, H., Vanel, F., Fuchida, S., Koshikawa, H., Yamagishi, T., Yamamoto, H. and Kawachi, M. 2020. Differential heavy metal sensitivity in seven algal species from the NIES culture collection based on delayed fluorescence assays. Phycol. Res. 68: 41–49.
- Petersen, S., Krätschell, A., Augustin, N., Jamieson, J., Hein, J. R. and Hannington, M. D. 2016. News from the seabed—Geological characteristics and resource potential of deep-sea mineral resources. Mar. Policy 70: 175–187.
- Roederer, M. 2008. How many events is enough? Are you positive? Cytometry A 73A: 384–385.
- Sawada, T., Watanabe, M., Fujimura, Y., Yagishita, S., Shimoyama, T., Maeda, Y., Kanda, S., Yunokawa, M., Tamura, K., Tamura, T., Minami, H., Koh, Y. and Koizumi, F. 2016. Sensitive cytometry based system for enumeration, capture and analysis of gene mutations of circulating tumor cells. Cancer Sci. 107: 307–314.
- Suman, T. Y., Radhika Rajasree, S. R. and Kirubagaran, R. 2015. Evaluation of zinc oxide nanoparticles toxicity on marine algae chlorella vulgaris through flow cytometric, cytotoxicity and oxidative stress analysis. Ecotoxicol. Environ. Saf. 113: 23–30.
- Thompson, A. W. and van den Engh, G. 2016. A multi-laser flow cytometry method to measure single cell and population-level relative fluorescence action spectra for the targeted study and isolation of phytoplankton in complex assemblages. Limnol. Oceanogr. Methods 14: 39–49.
- van den Engh, G., Ulloa, O. and Karl, D. M. 2019. Monitoring microbial communities in the marine environment. Cytometry A 95: 717–721.
- Van Dover, C. L. 2011. Mining seafloor massive sulphides and biodiversity: What is at risk? ICES J. Mar. Sci. 68: 341–348.
- Watanabe, M., Serizawa, M., Sawada, T., Takeda, K., Takahashi, T., Yamamoto, N., Koizumi, F. and Koh, Y. 2014. A novel flow cytometry-based cell capture platform for the detection, capture and molecular characterization of rare tumor cells in blood. J. Transl. Med. 12: 143.
- Wedding, L. M., Reiter, S. M., Smith, C. R., Gjerde, K. M., Kittinger, J. N., Friedlander, A. M., Gaines, S. D., Clark, M. R., Thurnherr, A. M., Hardy, S. M. and Crowder, L. B. 2015. Managing mining of the deep seabed: Contracts are being granted, but protections are lagging. Science 349: 144–145.
- Xiao, X., Han, Z. Y., Chen, Y. X., Liang, X. Q., Li, H. and Qian, Y. C. 2011. Optimization of FDA-PI method using flow cytometry to measure metabolic activity of the cyanobacteria, Microcystis aeruginosa. Phys. Chem. Earth 36: 424–429.
- Yamagishi, T., Fuchida, S., Katsumata, M., Horie, Y., Mori, F., Kitayama, A., Kawachi, M., Koshikawa, H., Nozaki, T., Kumagai, H., Ishibashi, J. and Tatarazako, N. 2018. Evaluation of the toxicity of leaches from hydrothermal sulfide deposits by means of a delayed fluorescence-based bioassay with the marine cyanobacterium Cyanobium sp. NIES-981. Ecotoxicology 27: 1303–1309.
- Yamagishi, T., Katsumata, M., Yamaguchi, H., Shimura, Y., Kawachi, M., Koshikawa, H., Horie, Y. and Tatarazako, N. 2016. Rapid ecotoxicological bioassay using delayed fluorescence in the marine cyanobacterium Cyanobium sp. (NIES-981). Ecotoxicology 25: 1751–1758.
- Yamaguchi, H., Shimura, Y., Suzuki, S., Yamagishi, T., Tatarazako, N. and Kawachi, M. 2016. Complete genome sequence of Cyanobium sp. NIES-981, a marine strain potentially useful for ecotoxicological bioassays. Genome Announc. 4: e00736–e16.
- Zhang, H., Zhang, S., Peng, Y., Li, Y., Chen, Z., Xu, H., Yu, Z., Zheng, W. and Zheng, T. 2015. Effects of marine actinomycete on the removal of a toxicity alga Phaeocystis globose in eutrophication waters. Front. Microbiol. 6: 474.
